---
title: "VulAgent — paper by Wang et al. (2026) — summary and ideas"
description: "Language-model vulnerability detectors often flag benign code. 36 references. Among them: 4."
authors: ["Ziliang Wang", "Ge Li", "Jia Li", "Hao Zhu", "Zhi Jin"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.928/"
doi: "10.18653/v1/2026.findings-acl.928"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/wang-2026-vulagent-hypothesis-validation-driven-multi/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wang-2026-vulagent-hypothesis-validation-driven-multi/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/4/) / VulAgent: Hypothesis-Validation Driven Multi-Agent Architecture for Vulnerability Detection

# VulAgent: Hypothesis-Validation Driven Multi-Agent Architecture for Vulnerability Detection

**Authors:** [Ziliang Wang](https://md.research.elamur.ai/free/en/authors/ziliang-wang/), [Ge Li](https://md.research.elamur.ai/free/en/authors/ge-li/), [Jia Li](https://md.research.elamur.ai/free/en/authors/jia-li/), [Hao Zhu](https://md.research.elamur.ai/free/en/authors/hao-zhu/) and [Zhi Jin](https://md.research.elamur.ai/free/en/authors/zhi-jin/) · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: ACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.findings-acl.928/) · [PDF](https://aclanthology.org/2026.findings-acl.928.pdf) · [DOI 10.18653/v1/2026.findings-acl.928](https://doi.org/10.18653/v1/2026.findings-acl.928) · [OpenAlex](https://openalex.org/W7166818667)

**Citations:** 4 citations (OpenAlex, 2026-10-14)

## Summary
Language-model vulnerability detectors often flag benign code. They notice a risky operation but cannot confirm whether surrounding guards actually block exploitation. VulAgent addresses this by modeling the workflow of human code auditors. Each suspicion becomes an explicit, falsifiable hypothesis that must pass evidence checks before it is reported.

The pipeline has three stages:
1. **Discovery.** A MetaAgent dispatches role-specialized analyzers. This stage aims for high recall and merges duplicate findings by CWE and sink line.
2. **Hypothesis construction.** A TriggerPlannerAgent states the preconditions for the vulnerability and a candidate source-to-sink trigger path.
3. **Validation.** The conditions and then the path are checked against project context and general knowledge of APIs and security patterns.

The approach was evaluated on PrimeVul and SVEN against recent LLM-based baselines. VulAgent raises accuracy and identifies far more vulnerable–fixed pairs. It also cuts false positives by roughly a third. The value is practical: separating discovery from validation makes reports more trustworthy and lowers triage effort, though it requires more inference tokens.

## Key points
- Measured against recent LLM baselines on the PrimeVul and SVEN benchmarks, VulAgent:
  - raises accuracy by an average of 6.6 percentage points;
  - finds up to 4.5× more vulnerable–fixed pairs (2.46× on average);
  - lowers false positive rate by about 36% relative.
- The architecture combines a MetaAgent with always-on StaticAnalyzerAgent, BehaviorAnalyzerAgent and MemoryLayoutAgent, plus specialized analyzers. It then uses a TriggerPlannerAgent and two validation steps: hypothesis conditions, then hypothesis path.
- In an experiment where oracle labels are supplied as candidate reports, VulAgent reaches an FPR of 20.48%. This compares with 49.40% for direct CoT judgment and 39.76% for VulTrial. Accuracy stays close to direct CoT (61.45 vs 62.05).
- The main cost is output length: about 7,037 output tokens per sample, versus about 3,119 for VulTrial and 2,264 for GPTLens.
- Code and resources are released at github.com/HotFrom/VulAgent.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Peking University](https://md.research.elamur.ai/free/en/universities/peking-university/) — University, Beijing, China · 244 works · 39,359 citations
- [Central South University](https://md.research.elamur.ai/free/en/universities/central-south-university/) — University, Changsha, China · 15 works · 582 citations

📚 Shelves (3): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🗄️ Information Systems](https://md.research.elamur.ai/free/en/shelves/computer-science/information-systems/) · 1,437 works · [💻 Software](https://md.research.elamur.ai/free/en/shelves/computer-science/software/) · 92 works

## Cite

Cite as: Ziliang Wang, Ge Li, Jia Li, Hao Zhu and Zhi Jin. 2026. VulAgent: Hypothesis-Validation Driven Multi-Agent Architecture for Vulnerability Detection. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{wang2026vulagent,
  title = {{VulAgent: Hypothesis-Validation Driven Multi-Agent Architecture for Vulnerability Detection}},
  author = {Ziliang Wang, Ge Li, Jia Li, Hao Zhu and Zhi Jin},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.928},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.928}
}
```

- [← previous VoiceStar: Robust Zero-Shot Autoregressive TTS with Duration Control and Extrapolation](https://md.research.elamur.ai/free/en/papers/peng-2026-voicestar-robust-zero-shot-autoregressive/)
- [next → What Makes a Good Curriculum? Disentangling the Effects of Data Ordering on LLM Mathematical Reasoning](https://md.research.elamur.ai/free/en/papers/jia-2026-what-makes-good-curriculum-disentangling/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/4/)

## References

36 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Josh Achiam, Steven Adler, Sandhini Agarwal, and et al. (OpenAI). 2023. [GPT-4 technical report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/). arXiv preprint arXiv:2303.08774.
2. Liqing Cao, Haofeng Li, Chenghang Shi, Jie Lu, Haining Meng, Lian Li, and Jingling Xue. 2025. Vulpa: Detecting semantically recurring vulnerabilities with multi-object typestate analysis. Proceedings of the ACM on Software Engineering, 2(FSE):2430–2453. [find](https://scholar.google.com/scholar?q=Liqing+Cao%2C+Haofeng+Li%2C+Chenghang+Shi%2C+Jie+Lu%2C+Haining+Meng%2C+Lian+Li%2C+and+Jingling+Xue%2E+2025%2E+Vulpa%3A+Detecting+semantically+recurring+vulnerabilities+with+multi%2Dobject+typestate+analysis%2E+Proceedings+)
3. Checkmarx. 2022. Online. [https://www.checkmarx.com/](https://www.checkmarx.com/). [↗](https://www.checkmarx.com/)
4. Available: Xiao Cheng, Guanqin Zhang, Haoyu Wang, and Yulei Sui. 2022. Path-sensitive code embedding via contrastive learning for software vulnerability detection. In Proceedings of the 31st ACM SIGSOFT International Symposium on Software Testing and Analysis, pages 519–531. [find](https://scholar.google.com/scholar?q=Available%3A+Xiao+Cheng%2C+Guanqin+Zhang%2C+Haoyu+Wang%2C+and+Yulei+Sui%2E+2022%2E+Path%2Dsensitive+code+embedding+via+contrastive+learning+for+software+vulnerability+detection%2E+In+Proceedings+of+the+31st+ACM+SIGSO)
5. Hoa Khanh Dam, Truyen Tran, Trang Pham, Shien Wee Ng, John Grundy, and Aditya Ghose. 2017. Automatic feature learning for vulnerability prediction. arXiv preprint arXiv:1708.02368. — [arXiv:1708.02368](https://arxiv.org/abs/1708.02368) [find](https://scholar.google.com/scholar?q=Hoa+Khanh+Dam%2C+Truyen+Tran%2C+Trang+Pham%2C+Shien+Wee+Ng%2C+John+Grundy%2C+and+Aditya+Ghose%2E+2017%2E+Automatic+feature+learning+for+vulnerability+prediction%2E+arXiv+preprint+arXiv%3A1708%2E02368%2E+%E2%80%94+%5BarXiv%3A1708%2E02368)
6. DeepSeek-AI. 2024. DeepSeek-V3 technical report. arXiv preprint arXiv:2412.19437. [find](https://scholar.google.com/scholar?q=DeepSeek%2DAI%2E+2024%2E+DeepSeek%2DV3+technical+report%2E+arXiv+preprint+arXiv%3A2412%2E19437%2E)
7. Yangruibo Ding, Yanjun Fu, Omniyyah Ibrahim, Chawin Sitawarin, Xinyun Chen, Basel Alomair, David Wagner, Baishakhi Ray, and Yizheng Chen. 2025. Vulnerability detection with code language models: How far are we? In 2025 IEEE/ACM 47th International Conference on Software Engineering (ICSE), pages 1729… [find](https://scholar.google.com/scholar?q=Yangruibo+Ding%2C+Yanjun+Fu%2C+Omniyyah+Ibrahim%2C+Chawin+Sitawarin%2C+Xinyun+Chen%2C+Basel+Alomair%2C+David+Wagner%2C+Baishakhi+Ray%2C+and+Yizheng+Chen%2E+2025%2E+Vulnerability+detection+with+code+language+models%3A+How+f)
8. Flawfinder. 2022. Online. [http://www.dwheeler.com](http://www.dwheeler.com). [↗](http://www.dwheeler.com)
9. Available: Michael Fu and Chakkrit Tantithamthavorn. 2022. Linevul: A transformer-based line-level vulnerability prediction. In Proceedings of the 19th International Conference on Mining Software Repositories, pages 608–620. [find](https://scholar.google.com/scholar?q=Available%3A+Michael+Fu+and+Chakkrit+Tantithamthavorn%2E+2022%2E+Linevul%3A+A+transformer%2Dbased+line%2Dlevel+vulnerability+prediction%2E+In+Proceedings+of+the+19th+International+Conference+on+Mining+Software+Repo)
10. Daya Guo, Shuai Lu, Nan Duan, Yanlin Wang, Ming Zhou, and Jian Yin. 2022. [Unixcoder: Unified cross-modal pre-training for code representation](https://md.research.elamur.ai/free/en/papers/guo-2022-unixcoder-unified-cross-modal-pre/). In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 7212–7225.
11. Jingxuan He and Martin Vechev. 2023. Large language models for code: Security hardening and adversarial testing. In Proceedings of the 2023 ACM SIGSAC Conference on Computer and Communications Security, pages 1865–1879. [find](https://scholar.google.com/scholar?q=Jingxuan+He+and+Martin+Vechev%2E+2023%2E+Large+language+models+for+code%3A+Security+hardening+and+adversarial+testing%2E+In+Proceedings+of+the+2023+ACM+SIGSAC+Conference+on+Computer+and+Communications+Securit)
12. Junda He, Christoph Treude, and David Lo. 2025. LLM-based multi-agent systems for software engineering: Literature review, vision, and the road ahead. ACM Transactions on Software Engineering and Methodology, 34(5):1–30. [find](https://scholar.google.com/scholar?q=Junda+He%2C+Christoph+Treude%2C+and+David+Lo%2E+2025%2E+LLM%2Dbased+multi%2Dagent+systems+for+software+engineering%3A+Literature+review%2C+vision%2C+and+the+road+ahead%2E+ACM+Transactions+on+Software+Engineering+and+Meth)
13. Sihao Hu, Tiansheng Huang, Fatih İlhan, Selim Furkan Tekin, and Ling Liu. 2023. Large language model-powered smart contract vulnerability detection: New perspectives. In 2023 5th IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS-ISA), pages 29… [find](https://scholar.google.com/scholar?q=Sihao+Hu%2C+Tiansheng+Huang%2C+Fatih+I%CC%87lhan%2C+Selim+Furkan+Tekin%2C+and+Ling+Liu%2E+2023%2E+Large+language+model%2Dpowered+smart+contract+vulnerability+detection%3A+New+perspectives%2E+In+2023+5th+IEEE+International+C)
14. Julian Jang-Jaccard and Surya Nepal. 2014. A survey of emerging threats in cybersecurity. Journal of computer and system sciences, 80(5):973–993. [find](https://scholar.google.com/scholar?q=Julian+Jang%2DJaccard+and+Surya+Nepal%2E+2014%2E+A+survey+of+emerging+threats+in+cybersecurity%2E+Journal+of+computer+and+system+sciences%2C+80%285%29%3A973%E2%80%93993%2E)
15. Arnold Johnson, Kelley Dempsey, Ron Ross, Sarbari Gupta, Dennis Bailey, et al. 2011. Guide for security-focused configuration management of information systems. NIST special publication, 800(128):16–16. [find](https://scholar.google.com/scholar?q=Arnold+Johnson%2C+Kelley+Dempsey%2C+Ron+Ross%2C+Sarbari+Gupta%2C+Dennis+Bailey%2C+et+al%2E+2011%2E+Guide+for+security%2Dfocused+configuration+management+of+information+systems%2E+NIST+special+publication%2C+800%28128%29%3A16%E2%80%931)
16. Yue Li, Xiao Li, Hao Wu, Minghui Xu, Yue Zhang, Xiuzhen Cheng, Fengyuan Xu, and Sheng Zhong. 2025. Everything you wanted to know about LLM-based vulnerability detection but were afraid to ask. arXiv preprint arXiv:2504.13474. — [arXiv:2504.13474](https://arxiv.org/abs/2504.13474) [find](https://scholar.google.com/scholar?q=Yue+Li%2C+Xiao+Li%2C+Hao+Wu%2C+Minghui+Xu%2C+Yue+Zhang%2C+Xiuzhen+Cheng%2C+Fengyuan+Xu%2C+and+Sheng+Zhong%2E+2025%2E+Everything+you+wanted+to+know+about+LLM%2Dbased+vulnerability+detection+but+were+afraid+to+ask%2E+arXiv+p)
17. Zhen Li, Deqing Zou, Shouhuai Xu, Hai Jin, Yawei Zhu, and Zhaoxuan Chen. 2021. Sysevr: A framework for using deep learning to detect software vulnerabilities. IEEE Transactions on Dependable and Secure Computing, 19(4):2244–2258. [find](https://scholar.google.com/scholar?q=Zhen+Li%2C+Deqing+Zou%2C+Shouhuai+Xu%2C+Hai+Jin%2C+Yawei+Zhu%2C+and+Zhaoxuan+Chen%2E+2021%2E+Sysevr%3A+A+framework+for+using+deep+learning+to+detect+software+vulnerabilities%2E+IEEE+Transactions+on+Dependable+and+Secur)
18. Shuai Lu, Daya Guo, Shuo Ren, Junjie Huang, Alexey Svyatkovskiy, Ambrosio Blanco, Colin Clement, Dawn Drain, Daxin Jiang, Duyu Tang, et al. 2021. Codexglue: A machine learning benchmark dataset for code understanding and generation. arXiv preprint arXiv:2102.04664. [find](https://scholar.google.com/scholar?q=Shuai+Lu%2C+Daya+Guo%2C+Shuo+Ren%2C+Junjie+Huang%2C+Alexey+Svyatkovskiy%2C+Ambrosio+Blanco%2C+Colin+Clement%2C+Dawn+Drain%2C+Daxin+Jiang%2C+Duyu+Tang%2C+et+al%2E+2021%2E+Codexglue%3A+A+machine+learning+benchmark+dataset+for+co)
19. Wei Ma, Daoyuan Wu, Yuqiang Sun, Tianwen Wang, Shangqing Liu, Jian Zhang, Yue Xue, and Yang Liu. 2024. Combining fine-tuning and LLM-based agents for intuitive smart contract auditing with justifications. arXiv preprint arXiv:2403.16073. — [arXiv:2403.16073](https://arxiv.org/abs/2403.16073) [find](https://scholar.google.com/scholar?q=Wei+Ma%2C+Daoyuan+Wu%2C+Yuqiang+Sun%2C+Tianwen+Wang%2C+Shangqing+Liu%2C+Jian+Zhang%2C+Yue+Xue%2C+and+Yang+Liu%2E+2024%2E+Combining+fine%2Dtuning+and+LLM%2Dbased+agents+for+intuitive+smart+contract+auditing+with+justificati)
20. Zhenyy Mao, Jialong Li, Dongming Jin, Munan Li, and Kenji Tei. 2024. Multi-role consensus through LLMs discussions for vulnerability detection. In 2024 IEEE 24th International Conference on Software Quality, Reliability, and Security Companion (QRS-C), pages 1318–1319. IEEE. [find](https://scholar.google.com/scholar?q=Zhenyy+Mao%2C+Jialong+Li%2C+Dongming+Jin%2C+Munan+Li%2C+and+Kenji+Tei%2E+2024%2E+Multi%2Drole+consensus+through+LLMs+discussions+for+vulnerability+detection%2E+In+2024+IEEE+24th+International+Conference+on+Software+Q)
21. Yu Nong, Mohammed Aldeen, Long Cheng, Hongxin Hu, Feng Chen, and Haipeng Cai. 2024. Chain-of-thought prompting of large language models for discovering and fixing software vulnerabilities. arXiv preprint arXiv:2402.17230. — [arXiv:2402.17230](https://arxiv.org/abs/2402.17230) [find](https://scholar.google.com/scholar?q=Yu+Nong%2C+Mohammed+Aldeen%2C+Long+Cheng%2C+Hongxin+Hu%2C+Feng+Chen%2C+and+Haipeng+Cai%2E+2024%2E+Chain%2Dof%2Dthought+prompting+of+large+language+models+for+discovering+and+fixing+software+vulnerabilities%2E+arXiv+prepr)
22. Qwen Team. 2025. [Qwen3 technical report](https://md.research.elamur.ai/free/en/papers/yang-2025-qwen3-technical-report/). arXiv preprint arXiv:2505.09388.
23. Rebecca Russell, Louis Kim, Lei Hamilton, Tomo Lazovich, Jacob Harer, Onur Ozdemir, Paul Ellingwood, and Marc McConley. 2018. Automated vulnerability detection in source code using deep representation learning. In 2018 17th IEEE international conference on machine learning and applications (ICMLA),… [find](https://scholar.google.com/scholar?q=Rebecca+Russell%2C+Louis+Kim%2C+Lei+Hamilton%2C+Tomo+Lazovich%2C+Jacob+Harer%2C+Onur+Ozdemir%2C+Paul+Ellingwood%2C+and+Marc+McConley%2E+2018%2E+Automated+vulnerability+detection+in+source+code+using+deep+representation)
24. Jingwei Shi, Xinxiang Yin, Jing Huang, Jinman Zhao, and Shengyu Tao. 2026. [Codehacker: Automated test case generation for detecting vulnerabilities in competitive programming solutions](https://md.research.elamur.ai/free/en/papers/shi-2026-codehacker-automated-test-case-generation/). arXiv preprint arXiv:2602.20213.
25. Benjamin Steenhoek, Hongyang Gao, and Wei Le. 2024a. Dataflow analysis-inspired deep learning for efficient vulnerability detection. In Proceedings of the 46th ieee/acm international conference on software engineering, pages 1–13. [find](https://scholar.google.com/scholar?q=Benjamin+Steenhoek%2C+Hongyang+Gao%2C+and+Wei+Le%2E+2024a%2E+Dataflow+analysis%2Dinspired+deep+learning+for+efficient+vulnerability+detection%2E+In+Proceedings+of+the+46th+ieee%2Facm+international+conference+on+sof)
26. Benjamin Steenhoek, Md Mahbubur Rahman, Monoshi Kumar Roy, Mirza Sanjida Alam, Earl T Barr, and Wei Le. 2024b. A comprehensive study of the capabilities of large language models for vulnerability detection. arXiv preprint arXiv:2403.17218, 59. — [arXiv:2403.17218](https://arxiv.org/abs/2403.17218) [find](https://scholar.google.com/scholar?q=Benjamin+Steenhoek%2C+Md+Mahbubur+Rahman%2C+Monoshi+Kumar+Roy%2C+Mirza+Sanjida+Alam%2C+Earl+T+Barr%2C+and+Wei+Le%2E+2024b%2E+A+comprehensive+study+of+the+capabilities+of+large+language+models+for+vulnerability+dete)
27. Karl Tamberg and Hayretdin Bahsi. 2025. Harnessing large language models for software vulnerability detection: A comprehensive benchmarking study. IEEE Access, 13:29698–29717. [find](https://scholar.google.com/scholar?q=Karl+Tamberg+and+Hayretdin+Bahsi%2E+2025%2E+Harnessing+large+language+models+for+software+vulnerability+detection%3A+A+comprehensive+benchmarking+study%2E+IEEE+Access%2C+13%3A29698%E2%80%9329717%2E)
28. Chandra Thapa, Seung Ick Jang, Muhammad Ejaz Ahmed, Seyit Camtepe, Josef Pieprzyk, and Surya Nepal. 2022. Transformer-based language models for software vulnerability detection. In Proceedings of the 38th Annual Computer Security Applications Conference, pages 481–496. [find](https://scholar.google.com/scholar?q=Chandra+Thapa%2C+Seung+Ick+Jang%2C+Muhammad+Ejaz+Ahmed%2C+Seyit+Camtepe%2C+Josef+Pieprzyk%2C+and+Surya+Nepal%2E+2022%2E+Transformer%2Dbased+language+models+for+software+vulnerability+detection%2E+In+Proceedings+of+the+)
29. Saad Ullah, Mingji Han, Saurabh Pujar, Hammond Pearce, Ayse Coskun, and Gianluca Stringhini. 2024. LLMs cannot reliably identify and reason about security vulnerabilities (yet?): A comprehensive evaluation, framework, and benchmarks. In 2024 IEEE symposium on security and privacy (SP), pages 862– 88… [find](https://scholar.google.com/scholar?q=Saad+Ullah%2C+Mingji+Han%2C+Saurabh+Pujar%2C+Hammond+Pearce%2C+Ayse+Coskun%2C+and+Gianluca+Stringhini%2E+2024%2E+LLMs+cannot+reliably+identify+and+reason+about+security+vulnerabilities+%28yet%3F%29%3A+A+comprehensive+evalu)
30. Zhiyuan Wei, Jing Sun, Zijiang Zhang, Xianhao Zhang, Meng Li, and Zhe Hou. 2024. LLM-smartaudit: Advanced smart contract vulnerability detection. arXiv preprint arXiv:2410.09381. — [arXiv:2410.09381](https://arxiv.org/abs/2410.09381) [find](https://scholar.google.com/scholar?q=Zhiyuan+Wei%2C+Jing+Sun%2C+Zijiang+Zhang%2C+Xianhao+Zhang%2C+Meng+Li%2C+and+Zhe+Hou%2E+2024%2E+LLM%2Dsmartaudit%3A+Advanced+smart+contract+vulnerability+detection%2E+arXiv+preprint+arXiv%3A2410%2E09381%2E+%E2%80%94+%5BarXiv%3A2410%2E09381%5D%28)
31. Xin-Cheng Wen, Yijun Yang, Cuiyun Gao, Yang Xiao, and Deheng Ye. 2025. Boosting vulnerability detection of LLMs via curriculum preference optimization with synthetic reasoning data. arXiv preprint arXiv:2506.07390. [find](https://scholar.google.com/scholar?q=Xin%2DCheng+Wen%2C+Yijun+Yang%2C+Cuiyun+Gao%2C+Yang+Xiao%2C+and+Deheng+Ye%2E+2025%2E+Boosting+vulnerability+detection+of+LLMs+via+curriculum+preference+optimization+with+synthetic+reasoning+data%2E+arXiv+preprint+arX)
32. Xin-Cheng Wen, Jiaxin Ye, Cuiyun Gao, Lianwei Wu, and Qing Liao. 2024. Evalsva: Multi-agent evaluators for next-gen software vulnerability assessment. arXiv preprint arXiv:2501.14737. — [arXiv:2501.14737](https://arxiv.org/abs/2501.14737) [find](https://scholar.google.com/scholar?q=Xin%2DCheng+Wen%2C+Jiaxin+Ye%2C+Cuiyun+Gao%2C+Lianwei+Wu%2C+and+Qing+Liao%2E+2024%2E+Evalsva%3A+Multi%2Dagent+evaluators+for+next%2Dgen+software+vulnerability+assessment%2E+arXiv+preprint+arXiv%3A2501%2E14737%2E+%E2%80%94+%5BarXiv%3A2501%2E14)
33. Fabian Yamaguchi. 2017. Pattern-based methods for vulnerability discovery. it-Information Technology, 59(2):101–106. [find](https://scholar.google.com/scholar?q=Fabian+Yamaguchi%2E+2017%2E+Pattern%2Dbased+methods+for+vulnerability+discovery%2E+it%2DInformation+Technology%2C+59%282%29%3A101%E2%80%93106%2E)
34. Chenyuan Zhang, Hao Liu, Jiutian Zeng, Kejing Yang, Yuhong Li, and Hui Li. 2024. Prompt-enhanced software vulnerability detection using chatgpt. In Proceedings of the 2024 IEEE/ACM 46th International Conference on Software Engineering: Companion Proceedings, pages 276–277. [find](https://scholar.google.com/scholar?q=Chenyuan+Zhang%2C+Hao+Liu%2C+Jiutian+Zeng%2C+Kejing+Yang%2C+Yuhong+Li%2C+and+Hui+Li%2E+2024%2E+Prompt%2Denhanced+software+vulnerability+detection+using+chatgpt%2E+In+Proceedings+of+the+2024+IEEE%2FACM+46th+International+)
35. Xin Zhou, Ting Zhang, and David Lo. 2024. Large language model for vulnerability detection: Emerging results and future directions. In Proceedings of the 2024 ACM/IEEE 44th International Conference on Software Engineering: New Ideas and Emerging Results, pages 47–51. [find](https://scholar.google.com/scholar?q=Xin+Zhou%2C+Ting+Zhang%2C+and+David+Lo%2E+2024%2E+Large+language+model+for+vulnerability+detection%3A+Emerging+results+and+future+directions%2E+In+Proceedings+of+the+2024+ACM%2FIEEE+44th+International+Conference+on)
36. Hao Zhu, Jia Li, Cuiyun Gao, Jiaru Qian, Yihong Dong, Huanyu Liu, Lecheng Wang, Ziliang Wang, Xiaolong Hu, and Ge Li. 2025. Specification-guided vulnerability detection with large language models. arXiv preprint arXiv:2511.04014. — [arXiv:2511.04014](https://arxiv.org/abs/2511.04014) [find](https://scholar.google.com/scholar?q=Hao+Zhu%2C+Jia+Li%2C+Cuiyun+Gao%2C+Jiaru+Qian%2C+Yihong+Dong%2C+Huanyu+Liu%2C+Lecheng+Wang%2C+Ziliang+Wang%2C+Xiaolong+Hu%2C+and+Ge+Li%2E+2025%2E+Specification%2Dguided+vulnerability+detection+with+large+language+models%2E+arX)

*Updated 31 October 2026*